Dactylo
Dactylo
Dactylo integrates common third-party speech-to-text algorithms to transcribe spoken recordings of just about any quality, from low-quality field recordings to high-quality broadcast recordings. Dactylo produces indexable text results of spoken words based on configurable language models with accuracy between 20 and 95 percent. Output is stored as XML files that can pass to a final MAM system or to mediARC. Available languages: Arabic, American and British English, French, German, Greek, Hebrew, Mandarin Chinese, Norwegian, Polish, Russian, Spanish.
actLINE MODULES
Technical Prerequisites
Requires Processorhost mediARC
Requires Processorhost jobDB
Compatible to WindowsServer 2005/2008, WinXP, Win7
Connection to Processorhost via TCP/IP
Products
